Vivian Meazza wrote:
> I have run several traces on fuel.nas, and I can see the
> /consumables/fuel/tank[0]/kill-when-empty being set, despite not
> appearing anywhere (I searched the entire directory) other than once
> in fuel.nas, and certainly not in my configuration file. Hence my
> original question.

That's just bizarre, and if true points to something really scary like
a memory corruption issue or reference goof in the interpreter and/or
property system.  Can you send the trace output?  Are you sure you can
do this repeatably?  If this is real, then the solution is
*ABSOLUTELY* not to hack around with Nasal code to make it work. :)

> I can also see outofFuel being set, and the engine being cut when
> tank[0] is empty and tank[1] has plenty of fuel in it.

Once again: this is not a bug.  By your original explanation (correct
me if I'm wrong): tank[1] is never selected, and simply feeds tank[0].
If it's not selected the engine won't feed from it.  If tank[0] is
empty, the engine will get no fuel.

Can you try being really, explicitly verbose about your problem?
Leave out the Nasal bugs and analysis.  Just give me the behavior you
want to see from the two tanks and two fuel selector switches.


Flightgear-devel mailing list

Reply via email to